Automotive Air Conditioning Information Forum (Archives)

Provided by

We've updated our forums!
Click here to visit the new forum

Archive Home

Search Auto AC Forum Archives

Running SD7B10 CCW

digidocs on Mon February 25, 2013 2:15 AM User is offline

Hi guys,

I've heard that running a Sanden SD7B10 CCW will be noisy. Is that noisy as in the wife won't like it, or noisy as in metal shavings?

The objective is a very light 100-130cc compressor that is CCW rotation. The SD7 is about 4.6 lbs lighter than the TM13. Another option would be a CCW TRS090 off of a Honda, but I can't find any weight specs or if it would be possible to retrofit a v-belt clutch to the TRS.

Thanks for the help---the level of knowledge on this forum is pretty impressive.


ice-n-tropics on Mon February 25, 2013 5:54 PM User is offline

The SD7B10 will run CCW w/ approx 75% durability. The failure mode is hammering movement of the fixed gear keyway in the aluminum block until the piston rod swedge into the planet plate hits the lower end of the cylinder bore.
Above about 2800 crpm the anti-rotation gears enter a backlash condition which causes progressively increasing vibration and noise. A discharge muffler is only 25% improvement in noise reduction. After market A/C manufacturers for Honda Civic (CCW engines) have used Sanden piston CW compressors w/ just upper rpm noise negatives. Pumping capacity (BTU) is the same CW or CCW. Your wife might wear her NRA shooting ear muffs. Double oil might reduce noise a little.

The anti-rotation orbiting gear is staked to the "planet plate" so that there is about 2.5 degrees advance timing so that at piston TDC the downstroke is initiated while there is the maximum number of spiral bevel orbiting gears in contact w/ the fixed gear, to avoid backlash in ther CW direction only. To properly make a SD7B10 run CCW, the fixed gear keyway machining must be rotated 5 degrees. So you could remachine the keyway at 185 degrees change in the block and 180 degrees different on the fixed gear for an ideal CCW piston wobble/orbiting plate compressor.

It's pretty sophisticated engineering to machine the PV grooves off of the Civic TR and shrink fit a Vee belt sleeve to the machined rotor pulley (Ref old AAC forum). If dia. isn't important, there may be a pulley swap that would work.
Hope this helps,

Isentropic Efficiency=Ratio of Theoretical Compression Energy/Actual Energy. How To Air Condition Your Hot Rod

Edited: Tue February 26, 2013 at 10:49 AM by ice-n-tropics

digidocs on Tue February 26, 2013 3:17 PM User is offline

That information is great. I think the noise would be tolerable unless it gets extremely loud (loud enough to be heard over an airplane propeller at takeoff --- hint, hint).

I'm struggling a bit on the appropriate compressor because the application parameters are somewhat challenging. At start up, compressor rpm is in the 2000-2400 rpm range. The cabin will be heat soaked and it is desirable to have a larger capacity to cool the environment quickly. The compressor can be in this rpm range for the first 5-30 minutes. Then the compressor accelerates to 5000-5800 rpm steady state for 5-60+ minutes.

The condenser is driven by electric fans and its capacity doesn't vary much with engine rpm. There may be some additional cooling from ram air, but the exact amount is difficult to quantify (and less than a standard automotive application). Of course it is desirable to keep the condenser as small as possible to because of packaging and weight constraints.

The situation I'm running into is that I can get pretty good performance in the 2200 rpm range with even a smaller compressor, however lots of bad things happen at the higher speed. Even though the compressor VE is plummeting (how much?), isentropic efficiency goes way down and the discharge pressures and temperatures go sky high ~400 psig with my little condenser. I've been using the Hysys chemical/physical process modeler to try to get an understanding of the performance under these conditions, but I have little/extrapolated data from the compressor and condenser manufacturers to feed the models. I'm definitely an amateur at this, but I have been scouring the net and reading Alexander Kargillis's design book. So far this forum is a really great source of practical experience to balance all of the theoretical material.

Things that are important
- light weight, less than 45lbs for entire system
- 12-16K BTU cooling capacity
- CCW rotation, v-belt drive (with a bunch of effort could be serpentine)
- avoiding too many cut-and-try iterations

Things that are not as important
- engine power/COP
- cost of components

At the moment the front runner configurations are:

Lower capacity, less risk
-1 ft^2 condenser, 500 cfm airflow at design point, 370 BTU/ft^2/degF capacity
-10-12K capacity depending on conditions

More capacity, more risk, lighter?
-SD7B10 -3.1 lbs
-larger condenser/bigger fan (still working on sizing) +X lbs
-12-16K capacity

Greetings from a fellow Texan,

ice-n-tropics on Tue February 26, 2013 5:46 PM User is offline

If you attended recent ASHRAE show in Dallas, there were light weight A/C flex hose & vane axial fans. Every one pound of A/C = one less pound fuel.
Not prudent to get your feet off the ground w/ CW comp running back-ass-wards at 5800 crpm continuous.
Better to get unmachined SD7B10 or SD5H09 fixed gear and machine the keyway 5 deg. corrected for CCW or other countermeasure. Cost should not be prohibitive.
Sanden Singapore has CCW SD5H14 in mass production w/ straight bevel gears. Detune/customize SD5H14 for optimum capacity.
Comp characteristics:
TR VE is nearly flat from 2000 to 6000 crpm.
TR IE barely decreases from 95% to 90% as rpm increases.
CCW TR mass flow could be restricted by suction pressure regulator, exceptionally low MOP on TXV & decrease evap CFM as Pd increases.
Component balance:
SD7B10 has enough VE, IE, HP graphs to manually interpolate a model or Japan may have a computer model developed. Mass flow calc's can be adjusted based on Sp absolute. I always adjust comp & heat X calc's 5% due to OCR (which manufacturers ignore).
Scroll and SDB body suction port comp like low SH of return gas. TR can handle extreme liquid slugging.
I have a pal who designs variable capacity scroll for Trane in Tyler.
Cond BTU rejection = comp HP + evap BTU absorbed + factor of safety.

Isentropic Efficiency=Ratio of Theoretical Compression Energy/Actual Energy. How To Air Condition Your Hot Rod

Edited: Tue February 26, 2013 at 5:54 PM by ice-n-tropics

digidocs on Tue February 26, 2013 7:28 PM User is offline

Hey Hod Rod, (sorry, couldn't resist.)

Thanks again for your reply. I missed the ASHRAE show because unfortunately no one pays me to tinker with this stuff.

Day job: Electrical Engineer
Night job: Building this in my garage ---

Hoses and fans:
I had planned to use Aeroquip GH134 series hose which is a reduced barrier design that can be "crimped" in-situ with special pliers. If there are newer lighter hoses out there, those would certainly be worth a look. For the condenser fan front, I have a couple in mind from Spal in the 2.2-5 lb range depending on how the air side dP turns out. Axial is small Spal single scroll for about 300 cfm at 1.2" dP. Again, any pointers to better technology is appreciated.

If you are allowed to share any tech data you have on the SD7B10, that would be wonderful. The data on the site is pretty minimal, US tech support claimed they have no VE curves, and I'm still waiting to hear from Japan. I like your idea of modifying the SD7B10, need to get a donor unit to take apart and study. Also, do you have any pointers to a commercial suction pressure regulator/valve? I had also considered compressor recirculation valve that could be opened at higher speeds.

Lastly, I feel like I am intruding a bit on your goodwill. Since it seems like you do this professionally, could I send you a little bit of beer money for your time? You can reach me directly at (my first name)

Take care,

Back to Automotive Air Conditioning Forum

We've updated our forums!
Click here to visit the new forum

Archive Home

Copyright © 2016 Arizona Mobile Air Inc.